首个能自动生成临床脑电报告的模型，让长时脑电数据变有用。
Neural Signals Generate Clinical Notes in the Wild
- 用预训练脑电与语言模型融合，实现多尺度端到端报告生成
- 在9048名患者1.1万小时数据上训练，表现全面超越现有方法
- 专家评估显示报告更符合临床判断，适合医疗AI研究者使用
从长时间、长度不一的脑电图（EEG）记录中生成总结异常模式、诊断发现和临床解释的临床报告仍需大量人工工作。我们提出CELM，首个能够对长时间、可变长度的脑电图进行摘要并实现多尺度端到端临床报告生成的临床脑电-语言基础模型。CELM通过整合预训练脑电基础模型与语言模型，实现可扩展的多模态学习。我们构建了一个大规模临床脑电数据集，包含来自9,048名患者的约11,000小时脑电记录，对应9,922份报告，并发布带有自动化报告结构化流程的基准数据集以促进后续研究。实验结果表明，CELM在所有评估设置下均持续优于现有方法。更重要的是，我们邀请临床专家进行人工评估，结果显示CELM生成的报告更具临床连贯性、诊断可靠性，并更贴近专家解读。模型与基准构建流程已开源：https://github.com/Jathurshan0330/CELM。
原文摘要 · Abstract (English)
Generating clinical reports that summarize abnormal patterns, diagnostic findings, and clinical interpretations from long-term EEG recordings remains labor-intensive. We present CELM, the first clinical EEG-to-Language foundation model capable of summarizing long-duration, variable-length EEG recordings and performing end-to-end clinical report generation at multiple scales. CELM integrates pretrained EEG foundation models with language models to enable scalable multimodal learning. We curate a large-scale clinical EEG dataset containing 9,922 reports paired with approximately 11,000 hours of EEG recordings from 9,048 patients to train CELM, and release the benchmark with an automated report-structuring pipeline to facilitate future research. Experimental results show that CELM consistently outperforms existing methods across all evaluation settings. Importantly, we further conduct human evaluation with clinical experts, demonstrating that CELM generates reports that are more clinically coherent, diagnostically reliable, and better aligned with expert interpretation. We release our model and benchmark construction pipeline at https://github.com/Jathurshan0330/CELM.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。